Famoxadone: The discovery and optimisation of a new agricultural fungicide

Article abstract of DOI:10.1002/1526-4998(200102)57:2<143::AID-PS282>3.0.CO;2-8

Famoxadone (3-anilino-5-methyl-5-(4-phenoxyphenyl)-1,3-oxazolidine-2,4-dione), is a new agricultural fungicide recently commercialized by DuPont under the trade name Famoxate. Famoxadone is a member of a new class of oxazolidinone fungicides th

Full text of DOI:10.1002/1526-4998(200102)57:2<143::AID-PS282>3.0.CO;2-8

Pest Management Science
Pest Manag Sci 57:143±152 (2001)
Famoxadone: the discovery and optimisation of
a new agricultural fungicide†
Jeffrey A Sternberg,1* Detlef Geffken,2 John B Adams Jr,1 Reiner Po¨stages,2
Charlene G Sternberg,1 Carlton L Campbell1 and William K Moberg1
1DuPont Crop Protection Products, EI du Pont de Nemours and Co, Stine-Haskell Research Center, Building 300, Newark, Delaware
19714, USA
2Institut Fu¨r Pharmazeutische Chemie, Universitat Hamburg, Bundesstrasse 45, 20146 Hamburg, Germany
Abstract: Famoxadone (3-anilino-5-methyl-5-(4-phenoxyphenyl)-1,3-oxazolidine-2,4-dione), is a new
agricultural fungicide recently commercialized by DuPont under the trade name Famoxate1.
Famoxadone is a member of a new class of oxazolidinone fungicides that demonstrate excellent
control of plant pathogens in the Ascomycete, Basidiomycete, and Oomycete classes that infect grapes,
cereals, tomatoes, potatoes and other crops. DuPont's entry into the oxazolidinone area resulted from
the procurement of 5-methyl-5-phenyl-3-phenylamino-2-thioxo-4-oxazolidinone (1) from Professor
Detlef Geffken, then at the University of Bonn. An extensive analog program was initiated immediately
after the fungicidal activity of 1 was discovered through routine greenhouse testing. The discovery
program in the oxazolidinone area eventually culminated in the advancement of famoxadone to
commercial development in the early 1990s. The synthesis of various oxazolidinone ring systems and
the development of the structure-activity relationships that led to the discovery of famoxadone are
described.
# 2001 Society of Chemical Industry
Keywords: oxazolidinone; oxazolidinedione; famoxadone; discovery; fungicide; chemistry
1
INTRODUCTION
The discovery of ef®cacious and environmentally
benign agrochemicals is becoming increasingly more
challenging as business and regulatory pressures
continue to rise. A discovery strategy that has proven
successful for DuPont, as well as for other agricultural
chemical companies, is the procurement and screening
of novel compounds from academic and industrial
sources around the world. This strategy has provided a
large and diverse array of novel compounds at a
relatively low cost.
Figure 1. Lead compound from D Geffken (1) and famoxadone.
compound was extremely intriguing. Vigorous chemi-
cal optimization programs ensued in both the DuPont
and Geffken laboratories, and new syntheses were
developed to facilitate the preparation of closely
related analogs, as well as to modify the core
oxazolidinone ring. The mode of action of the
oxazolidinones was determined to be inhibition of
mitochondrial electron transport at complex III. In
vitro assays were developed which supplemented data
provided from greenhouse and ®eld evaluations, and
they greatly enhanced the optimization effort. After
three years of work and the preparation of over 700
analogs, famoxadone was advanced to commercial
development.1±4 Not only was the discovery program
successful in ultimately bringing a product to market,
DuPont's entry into the oxazolidinone area resulted
from the procurement of 5-methyl-5-phenyl-3-phenyl-
amino-2-thioxo-4-oxazolidinone (Fig 1; 1) from Pro-
fessor Detlef Geffken, then at the University of Bonn,
Germany. Greenhouse testing of 1 in fungicide screens
showed 100% preventive control of tomato late blight
(Phytophthora infestans (Mont) de Bary) at 200mg
litre 1, and 97% preventive control of grape downy
mildew (Plasmopara viticola Berl and de Toni) at
10mg litre 1. Preventive control of these Oomycete
pathogens was a key fungicide product concept for
DuPont at the time, and the potency of the lead
* Correspondence to: Jeffrey A Sternberg, DuPont Crop Protection Products, EI du Pont de Nemours and Co, Stine-Haskell Research
Center, Building 300, PO Box 30, Newark, Delaware, 19714, USA
E-mail: jeffrey.a.sternberg@usa.dupont.com
† Based on a paper presented at the symposium ‘New Chemistries for Crop Protection’, organised by Dr L Copping and the late Dr J Dingwall
on behalf of the Bioactive Sciences and Crop Protection Groups of the SCI and held at 14/15 Belgrave Square, London, UK on 19 June 2000
(Received 19 June 2000; revised version received 27 September 2000; accepted 10 October 2000)
# 2001 Society of Chemical Industry. Pest Manag Sci 1526±498X/2001/$30.00
143

JA Sternberg et al
it also exempli®ed a true collaboration between
DuPont and the laboratory of Professor Geffken.
percentage disease estimates for the treated plants with
those of the untreated controls. On average, 98% and
85% of disease were present on untreated control
tomato and grape plants, respectively.
2
EXPERIMENTAL METHODS
Percentage disease control data in Tables 2±6 and 8
are mean values from multiple tests conducted during
the course of this work. Each test was performed with
three replicates. The average standard error measure-
ment (SEM) percentage values for the tests performed
during this period are listed in Table 1.
2.1 Chemicals
Most chemical starting materials were obtained from
Aldrich Chemical (Milwaukee, WI, USA), Sigma
Chemical (St Louis, MO, USA), VWR (Trenton,
NJ, USA) or Fluka Chemical (Buchs, Switzerland).
2.2 Chemical methods
Melting points were determined on a Thomas Hoover
Unimelt capillary melting point apparatus. [1H]NMR
spectra were recorded on a Varian EM 390 (90MHz)
or a Varian XL-200 spectrometer (200MHz) using
tetramethylsilane as an internal standard, in either
deuterochloroform or deuterodimethyl sulfoxide solu-
tion (chemical shifts in dppm). IR spectra were
recorded on a Mettler FP61 or a Perkin Elmer 1600
series FTIR. Thin-layer chromatograms were devel-
oped on glass pre-coated with silica gel F254. The spots
were revealed under UV light or with iodine.
2.4 Mitochondrial preparations
All buffers and puri®cation procedures described in
this section were at 0±4°C.
2.4.1 Rat heart mitochondria
Rat hearts were isolated using an established proce-
dure at Haskell Laboratory (Newark, DE, USA).
Immediately after extraction the hearts were placed in
sodium chloride solution (20g litre 1) and then
washed with buffer A, which contained sucrose
(250mM), Hepes±sodium hydroxide (20mM) and
EDTA (0.1mM; pH 7.5). The hearts were minced
with a razor blade and homogenized in buffer A (1‡4,
w/v) using a Dounce homogenizer with a loosely ®tting
pestle, followed by a ground-glass pestle. The homo-
genate was centrifuged for 10min at 900g. The
supernatant was collected and centrifuged for 40min
2.2.1 Synthesis of famoxadone
A
mixture of ethyl 2-(4-phenoxyphenyl)lactate
(0.052mol), 1,1'-carbonyldiimidazole (0.06mol) and
dry dichloromethane (100ml) was agitated at 25°C for
19h. Water (30ml) was then added and the mixture
was agitated for 15min. Acetic acid (5ml) and
phenylhydrazine (0.069mol) were then added, and
the mixture agitated at 25°C for 24h, after which
water (100ml) was added followed by hydrochloric
acid to lower the pH to 2; the aqueous layer was then
removed. After washing the dichloromethane layer
with water (50ml), the solvent was removed under
vacuum and the oily residue was mixed with hexane
(150ml) and ethyl acetate (15ml), heated to 65°C,
cooled to 20°C, and then ®ltered. The solids were
washed with a mixture of ethyl acetate (20ml) and
hexane (80ml) and then dried to give the product
(0.041mol). Mp 140±142°C; [1H]NMR (CDCl3) d:
7.55 (m, 2H); 7.40±7.10 (m, 8H); 7.07 (m, 4H); 6.75
(d, 2H); 6.07 (s, 1H); 1.99 (s, 3H).
at 30000g. The resulting submitochondrial pellet was
1
homogenized in buffer A to 20mg of proteinml
,
¯ash frozen in liquid nitrogen and stored at 80°C
until used.
2.4.2 Phytophthora mitochondria
Ph infestans sporangia were isolated by washing V8
agar plates with cold water and ®ltering the washings
through glass wool. Sporangia were incubated at 12°C
for 1±4h and were observed under a microscope to
determine when zoospores were being released. When
many zoospores were observed, aliquots of culture
medium were mixed with buffer A (1‡1 by volume).
Following two 30-s bursts in a mini-bead beater, the
lysate was centrifuged for 10min at 600g. The
supernatant was centrifuged for 20min at 20000g
and the mitochondrial pellet was suspended in a
minimal volume of buffer A. Mitochondria were used
for inhibition studies within 1±2h of isolation.
2.3 Greenhouse testing
Aqueous suspensions of test compounds were sprayed
onto grape or tomato seedlings to the point of run-off
and afterwards the plants were placed in a drying
chamber. The following day the seedlings were
inoculated with a spore suspension of P viticola, (the
causal agent of grape downy mildew) or Ph infestans
(the causal agent of tomato late blight) and then
incubated in a water-saturated atmosphere for 24h at
20°C. The plants were transferred to another growth
chamber for 6 days of incubation at 20°C. Grape
plants were then incubated for 24h in a water-
saturated atmosphere at 20°C, after which disease
ratings were made on both grape and tomato plants.
Percentage control was calculated by comparing the
Table 1. Average standard error measurement (SEM)
percentage values for disease control data
SEM (Æ) for
Data mean range Ph infestans P viticola
0±10
45±55
75±85
85±95
95±100
5.1
16.7
9.2
9.8
17.3
10.4
6.9
5.8
1.8
1.4
144
Pest Manag Sci 57:143±152 (2001)

Famoxadone
2.5 Determination of enzyme inhibition
3.2 Synthesis
3.2.1 Synthesis routes to oxazolidinones
Reaction mixtures (1ml) were held in quartz cuvettes
containing potassium phosphate (0.1M) and cyto-
chrome c (10mM) at pH 7.5 at 25°C. Submitochondria
(5±20mg of protein) was added to the reaction mixture
10min before the addition of inhibitor ethanol (5ml)
which was immediately followed by the addition of
NADH (100nmol) in ethanol (10ml). The reaction
(NADH to O2) was monitored continuously for 1min
at 340nm using an HP 8542A diode array spectro-
photometer. Linear initial rates were ®tted to a line
using the spectrophotometer's software. Inhibition
constants were determined by ®tting rates to eqn (1),
where n is the observed initial rate, V is the uninhibited
rate, I is the inhibitor concentration and IC50 is the
concentration of inhibitor which gives 50% inhibition.
The synthesis of various oxazolidinone ring systems
had already been reported in the literature when
DuPont became interested in this class of
chemistry.7±14 The syntheses used to prepare 3-
amino-oxazolidinones are illustrated in Fig 2.10,11
Treatment of N-methyl-2-hydroxyhydroxamic acid 2
with 1,1'-carbonyldiimidazole (CDI) afforded the
dioxazinedione 3a. Cyclization could also be accom-
plished using phosgene and triethylamine, albeit in
lower yield. Reaction of the hydroxamic acid 2 with
1,1'-thiocarbonyldiimidazole (TCDI) gave the corre-
sponding thioxodioxazinone 3b. The dioxazinedione
(3a or 3b) was then treated with a monosubstituted
hydrazine to give the 3-amino-oxazolidinone (4a or
4b, respectively). The ®ve-membered ring structure
was con®rmed by [1H]NMR and IR analyses, and
none of the isomeric six-membered ring oxadiazinones
were observed.15,16
V
n ˆ
ꢀ1†
1 ‡ I=IC50
These syntheses were quite general and afforded
appropriately high yields of oxazolidinones, but the
multi-step preparation of the starting hydroxamic
acids limited our progress. A shorter route was
envisaged involving treatment of 2-hydroxy hydrazides
with a carbonylating agent, as illustrated in Fig 3. The
glycolic acid hydrazide 5a reacted smoothly with
TCDI to form the ®ve-membered ring 2-thioxo-4-
oxazolidinone 6a in good yield. The reaction of
mandelic acid hydrazide 5b with TCDI did not
proceed as cleanly and a number of products were
formed. However, the reaction with CDI produced
mainly the desired 2,4-oxazolidinedione 6b. Hydra-
zides bearing a tertiary alcohol, such as 5c, reacted in
an entirely different manner. Treatment with TCDI
afforded the oxadiazol-2(3H)-one 7 in high yield
without any of the desired 2-thioxo-4-oxazolidinone
1. Apparently, in this case, acylation of the aniline
nitrogen proceeded faster than acylation of the
hydroxyl group, and subsequent cyclization led to
the ®ve-membered ring oxadiazol-2(3H)-one.17 This
synthetic limitation was a signi®cant set-back, since
the 5,5-disubstituted compounds had signi®cantly
3
RESULTS AND DISCUSSION
3.1 Mode of action
Famoxadone is a potent inhibitor of mitochondrial
electron transport blocking the function of ubiqui-
nol:cytochrome c oxidoreductase (bc1, complex III).5,6
Inhibitors at this site have been classi®ed into three
groups according to their interactions with the bc1
cytochrome. Oxazolidinones belong to group I, whose
other members include myxothiazol, strobilurins and
stigmatellins. Whereas all of these materials bind in the
Qo domain of the enzyme complex, oxazolidinones are
chemically distinct and represent a new class within
group I.
Biological effects observed for oxazolidinones were
consistent with inhibition of mitochondrial function.
When zoospores of the plant pathogen Ph infestans
were treated with famoxadone, oxygen consumption
ceased and the zoospores lost motility within seconds.6
Within minutes the zoospores disintegrated.6 When
mycelia of Ph infestans were treated with an oxazoli-
dinone, protein and nucleic acid biosyntheses were
curtailed.6
Figure 2. Literature syntheses of 3-amino-
oxazolidinones.
Pest Manag Sci 57:143±152 (2001)
145

JA Sternberg et al
Figure 5. Conversion of thioxo-oxazolidinones to oxazolidinediones.
oxymonosulfate (Oxone1) or aqueous silver nitrate7,18
(Fig 5). Not only did this procedure provide a route to
2,4-oxazolidinediones in high yield, but it also
generated both the oxo- and thioxo-compounds for
biological evaluation from a single synthetic sequence.
Based on considerations of biological activity, cost
and waste generation, interest shifted from the 2-
thioxo-4-oxazolidinones to the 2,4-oxazolidinediones.
It therefore became desirable to develop a synthesis
that provided these compounds directly, and avoided
the inef®ciency of introducing a sulfur atom and then
subsequently removing it. Biological results also
suggested that the 5-alkyl-5-aryl-3-phenylamino-oxa-
zolidinediones were the most desirable for maximum
fungicidal activity, and therefore efforts were directed
towards compounds with this substitution pattern.
A phosgene-based route was attempted ®rst (Fig 6).
Treatment of the 2-hydroxy ester 8 with phosgene and
N,N-diethylaniline (DEA) with a catalytic amount of
pyridine produced the chloroformate 9. The chlor-
oformate was not puri®ed but treated directly with
phenylhydrazine and N,N-diisopropylethylamine. Cy-
clization of the resulting carbazate was slow at room
temperature, and therefore the reaction was heated to
re¯ux in the presence of triethylamine (TEA). Yields
were generally low, several by-products were observed
(eg the acrylate 10) and unreacted starting material
was generally recovered despite efforts to drive the
phosgene reaction to completion. We now believe that
starting material was in fact consumed, and the
intermediate chloroformate 9 decomposed to the
tertiary chloride under the reaction conditions; the
tertiary chloride then hydrolyzed back to the starting
2-hydroxy ester during the aqueous work-up.
Figure 3. 3-Phenylamino-oxazolidinones from 2-hydroxy hydrazides.
higher fungicidal activity and, therefore, an alternative
synthetic route was pursued.
An ef®cient `one-pot' synthesis of 5,5-disubstituted
2-thioxo-4-oxazolidinones was developed starting
from readily available 2-hydroxy esters (Fig 4).
Sequential treatment of the ester 8 with potassium
tert-butoxide, carbon disul®de, ethyl chloroformate
and ®nally phenylhydrazine afforded the 2-thioxo-4-
oxazolidinone in good to excellent yield. The proce-
dure was fairly general and the product could often be
puri®ed without chromatography. The only disubsti-
tuted 2-hydroxy esters which worked poorly in this
reaction were those bearing strongly electron-with-
drawing groups (eg R1=CF3), powerful electron-
donating groups (eg R2=4-(CH3)2N±Ph), or those
with sterically hindered hydroxyl groups (eg R2=2,5-
(CH3)2±Ph).
Attempts to prepare the corresponding oxazolidine-
diones using the same methodology were unsuccess-
ful. Treatment of the 2-hydroxy ester 8 with a base,
carbonyl sul®de instead of carbon disul®de and,
®nally, an S-alkylating agent, failed to afford the
thiolcarbamate. Carbonyl sul®de was apparently
insuf®ciently electrophilic to react with the tertiary
alkoxide.
The need for a higher-yielding route to 2,4-
oxazolidinediones led to the development of the route
illustrated in Fig 7. Treatment of the 2-hydroxy ester 8
with CDI formed the acylimidazole 11, which was not
isolated but treated directly with phenylhydrazine and
acetic acid (HOAc) to give the 2,4-oxazolidinedione.
In general, yields were good and the products did not
require chromatography for puri®cation. Acetic acid
appeared to be critical to the success of the reaction
because without it no product was observed. It was
postulated that acetic acid protonated the imidazolyl
ring and thereby activated the carbonyl to nucleophilic
attack. Stronger acids, such as hydrochloric acid, gave
no reaction, presumably due to irreversible protona-
tion of the phenylhydrazine, rendering it non-nucleo-
philic.
2,4-Oxazolidinediones could, however, be obtained
in high yield by hydrolysis of the corresponding 2-
thioxo-4-oxazolidinone with aqueous potassium per-
With a general, direct and high-yielding route to
oxazolidinediones now available, a variety of R1 and R2
substituents could be surveyed quickly and easily. A
Figure 4. One-pot synthesis of 5,5-disubstituted 2-thioxo-4-
oxazolidinones.
146
Pest Manag Sci 57:143±152 (2001)

Famoxadone
Figure 6. Preparation of 2,4-oxazolidinediones
using phosgene.
Figure 7. Preparation of 2,4-oxazolidinediones
using 1,1'-carbonyldiimidazole (CDI).
wide assortment of 2-hydroxy esters was all that was
required. As illustrated in Fig 8, 2-hydroxy esters are
common building blocks and readily accessible by a
number of synthetic routes.19
testing indicated that most of the fungicidal activity
resided in the (S)-enantiomer. Later in the discovery
program, multigram quantities of (S)-famoxadone
were frequently required and a more ef®cient synthetic
sequence was needed. A route incorporating classical
resolution of 2-(4-phenoxy-phenyl)-lactic acid (12)
proved to be the most expeditious (Fig 9). A variety of
commercial homochiral amines was tested and quini-
dine found to be optimal in terms of selectivity and
crystallinity of only one of the diasteromeric salt
products. Treatment of the 2-hydroxy acid 12 with
0.55 equivalents of quinidine followed by separation,
esteri®cation and cyclization with CDI and phenylhy-
drazine afforded (R)- and (S)-famoxadone.
3.2.2 Chiral oxazolidinones
The majority of compounds prepared during the
course of the optimization program were racemates
possessing a single stereogenic center at C-5 of the
oxazolidinone ring. An obvious early question was
whether some, or all, of the biological activity resided
in one of the two enantiomers. Milligram quantities of
the two enantiomers of the original 2-thioxo-4-
oxazolidinone (1) were obtained and greenhouse
Figure 8. Methods for preparing 2-(substituted
phenyl)lactate esters. aTMSCN=trimethylsilyl
cyanide. bMCF=methyl chloroformate.
Pest Manag Sci 57:143±152 (2001)
147

JA Sternberg et al
Figure 9. Preparation of (R)- and (S)-
famoxadone by classical resolution.
Figure 10. Ring analogs of 2-thioxo-4-
oxazolidinone and 2,4-oxazolidinediones.
3.2.3 Ring analogs
one modi®cation that is discussed in greater detail
below is that of the 1-position replacement analogs
(13).
Variations of the central 2-thioxo-4-oxazolidinone or
2,4-oxazolidinedione ring were also investigated as
part of the optimization program (Fig 10).20 Modi®-
cations included replacement of oxygen at the 1-
position with sulfur, nitrogen or carbon; preparation of
2-imino analogs such as imines (=NR), oximes
(=NOR) and hydrazones (=NNR2),21 formation of
4-thiono (=S) and 4-imino (=NR) derivatives,22 and
the synthesis of two 6-member ring analogs. The
chemistry and biological activity of most of these
analogs will be the subject of future publications. The
Figure 11 illustrates the preparation of the sulfur
analog 13a. Treatment of methyl atrolactate (14) with
sulfonyl chloride and catalytic DMF afforded the
tertiary chloride 15 with a large amount of the
corresponding acrylate as a by-product. The two were
not separated but the mixture was treated with the
phenyl thiolcarbazate salt 16 and heated to give the
thio analog 13a. However, the yield was low and a
large amount of the methyl thiol carbazate 17 was also
Figure 11. Synthesis of ring analog 13a with sulfur
at the 1-position instead of oxygen.
148
Pest Manag Sci 57:143±152 (2001)

Famoxadone
the open-chain hydrazide, which was then cyclized
with heat to give the succinimide 13c.
3.3 Structure–activity relationships
3.3.1 Effect of various substituents on the 2,4-
oxazolidinedione ring
Over 700 analogs were prepared in the oxazolidinone
area. The following structure-activity relationships
were established based on testing in greenhouse
fungicide screens.
The data in Table 2 illustrate the structure-activity
relationships for R1, which is methyl in compound 1,
and clearly indicates that steric bulk has a signi®cant
effect on biological activity. For example, changing R1
from methyl to ethyl reduced preventive control of Ph
infestans and P viticola eightfold and fourfold, respec-
tively (1 vs 25). The R1=n-butyl compound 27 showed
no activity against these diseases at the rates indicated.
Reducing the size of R1 from methyl (1) to hydrogen
(24) also lowered activity. These data and others
suggested that a methyl group was the optimal
substituent at this position.
Figure 12. Preparation of hydantoin ring analog 13b.
isolated. As expected, the carbonyl stretching frequen-
cies in 13a were lower than in the oxazolidinedione,
due to the change in ring geometry caused by the larger
sulfur atom.
The 1-NH ring analog 13b was also prepared as
shown in Fig 12. Treatment of the amino ester 18 with
triphosgene afforded the isocyanate 19. Subsequent
condensation with phenylhydrazine afforded the N-
anilinourea 20, which was then cyclized with acetic
acid and heat to give the desired hydantoin 13b. An
attempt to methylate 13b with methyl tri¯uoromethyl
sulfonate failed to give either the N- or O-methyl
derivative. The 1-NCH3 derivative of the hydantoin
derived from the mandelate ester was ultimately
prepared (ie the analog lacking the 5-methyl group)
by a different chemical route.
Table 2. Effect of various R1 groups on preventive control of Phytophthora
infestans on tomatoes and Plasmopara viticola on grapes
The ®nal analog prepared in this series was the N-
anilino succinimide 13c wherein the 1-oxygen was
replaced with CH2 (Fig 13). Condensation of aceto-
phenone with ethyl cyanoacetate afforded the cinna-
mate 21 as a mixture of (E)- and (Z)-isomers.
Conjugate addition of cyanide followed by decarbox-
ylation afforded the nitrile 22, which was then
saponi®ed with potassium hydroxide to afford the
diacid 23. Activation of the acid with acetyl chloride
followed by reaction with phenyl hydrazine afforded
Disease control (%)
Compound
R1
Ph infestansa
P viticolab
1
24
25
26
27
CH3
H
100
86
33
0
97
8
CH3CH2
23
41
0
CH2 =CHCH2
CH3(CH2)2CH2
0
a
1
200mg litre
.
b
1
10mg litre
.
Figure 13. Synthesis of succinimide ring analog
13c.
Pest Manag Sci 57:143±152 (2001)
149

JA Sternberg et al
Table 3. Effect of various R3 groups on preventive control of Phytophthora
Table 5. Effect of R2 on preventive control of Phytophthora infestans on
infestans on tomatoes and Plasmopara viticola on grapes
tomatoes and Plasmopara viticola on grapes
Disease control (%)
Disease control (%)
R3
Ph infestansa
P viticolab
Compound
R2
Ph infestansa
P viticolab
Compound
44
45
46
47
48
49
50
51
52
53
54
55
CH3(CH2)4CH2
Cyclohexyl
Ph
100
48b
88
94
29
1
28
29
30
31
32
33
34
35
36
37
38
39
40
PhNH
Ph
100
0
100
15
48
PhCH2
PhCH2NH
0
43
2-F-Ph
100
17
100c
100
42
100
44a
97
15
0
5
2-CH3-Ph
3-Br-Ph
(Cyclohexyl)NH
(2-Pyridyl)NH
(2-CH3-Ph)NH
(2-Cl-Ph)NH
65
25
92
38
90
100
94
97
0
68
3-Cl-Ph
86
100
99
3-CH3-Ph
3-CF3O-Ph
4-F-Ph
8
60
6
(3-CH3-Ph)NH
(3-F-Ph)NH
91
93
56
100
97
4-Cl-Ph
76
17
(3-CH3O-Ph)NH
(4-F-Ph)NH
4-CH3O-Ph
4-CF3O-Ph
4-PhO-Ph
2,4-diF-Ph
15
25
100
73
56
Famoxadone
97
100
100
100
(4-CH3O-Ph)NH
(4-CF3-Ph)NH
95
0
89
57
100
a 200mg litre
b 40mg litre
.
1
a
b
c
1
1
200mg litre
40mg litre
.
.
1
.
Data for Ph infestans on potatoes.
The effect of varying R3, the group attached to the
oxazolidinone ring nitrogen atom which is NHPh in 1,
was also investigated (Table 3). The data for
compounds 28±32 suggest that an anilino or substi-
tuted anilino moiety at this position is critical for good
activity. Two oxazolidinones with R3=OPh were also
prepared and these too were signi®cantly less active,
especially against Ph infestans. Data for compounds
33±40 indicate that fungicidal activity generally de-
creased, sometimes dramatically, when substituents of
any type were introduced onto the phenyl ring. The
only substituted compounds that appeared to have
comparable activity were the 2-CH3, 3-F and 4-F
compounds 33, 36 and 38; however, additional testing
at lower concentrations showed that they too were less
active than 1. As with R1, the optimal substituent was
the one in the lead compound (1).
Somewhat surprisingly, a hydrogen atom on the
exocyclic nitrogen was not essential for fungicidal
activity. As indicated in Table 4, the N-methyl
compound 41 was also active, although less so than
1. The N-ethyl and N-acetyl compounds, 42 and 43
respectively, were signi®cantly less active, as were
oxazolidinones with larger N-alkyl groups. Once again,
the substituent in the lead compound 1 (R3=H) was
best at this position.
A diverse group of substituents could be tolerated at
the R2 position, as indicated in Table 5.23,24 Aryl
groups were preferred but the alkyl-substituted com-
pounds 44 and 45 were also active. In general, ortho-
substitution on the R2 phenyl ring was detrimental to
activity (eg see 48). Only a ¯uorine atom at this
position improved fungicidal control (see 47). Oxazo-
lidinones with a bromine or chlorine atom at the 3-
position were more active than the unsubstituted
compound (49, 50 vs 46), but other meta-substituted
compounds were generally less active. A wide variety
of substituents could be tolerated at the para-position
and many groups improved activity. Two of the most
active compounds were the 2,4-di¯uoro- and 4-
phenoxyoxazolidinones, 57 and famoxadone, respec-
tively. Famoxadone was preferred due to its lower
estimated manufacturing cost and superior residual
control properties.
Table 4. Effect of R4 on preventive control of Phytophthora
infestans on tomatoes and Plasmopara viticola on grapes
Disease control (%)
Compound
R4
Ph infestansa P viticolab
1
41
42
43
H
100
65
100
82
CH3
CH3CH2
CH3C(=O)
24
NTc
23
45
a
1
.
200mg litre
b
c
1
Once it was determined that a 4-phenoxy substi-
tuent was desirable, a wide variety of substituted-
40mg litre
.
NT=Not tested.
150
Pest Manag Sci 57:143±152 (2001)

Famoxadone
Table 6. Effect of R5 on preventive control of Phytopthora
Table 8. Activity of 1-oxygen ring analogs 13 for
infestans on tomatoes and Plasmopara viticola on grapes
preventive control of Phytophthora infestans on
tomatoes and Plasmopara viticola on grapes
% Disease control
Compound
R5
Ph infestansa P viticolab
% Disease control
58
59
60
61
62
63
64
65
66
67
68
69
2-i-Pr
2-Cl
0
98
100
98
92
64
74
97
97
68
0
84
68
Compound
Y
Ph infestansa P viticolab
46
O
S
88
15
48
0
2-F
82
13a
13b
13c
3-F
97
NH
CH2
77
35c
0
3-Cl
100
100c
100d
74
0
3-PhO
3-t-Bu
4-Cl
a 200mg litre
b 40mg litre
c Data for Ph infestans on potatoes.
.
1
1
.
4-Br
77
4-CH3O
4-Ph
4-t-Bu
99
96a,b,c
0
21
famoxadone were tested for intrinsic activity using in
vitro assays comprised of submitochondrial particles
isolated from rat heart and Ph infestans (Table 7). In
both compounds, the (S)-enantiomer was superior to
the (R)-enantiomer for inhibition of mitochondrial
electron transport, and (S)-famoxadone was approxi-
mately tenfold more potent than (S)-1. As one might
expect, the racemic mixtures were approximately half
as active as the (S)-enantiomers, and the (R)-
enantiomers were signi®cantly less active. The green-
house fungicide data on Ph infestans and P viticola were
consistent with the mitochondria data. The (S)-
enantiomers were the more active isomers, and the
racemic mixtures were approximately half as active.
The future commercialization of (S)-famoxadone as a
potential renewal strategy for famoxadone is being
considered.
a 200mg litre
.
1
b 2mg litre
c 5mg litre
d 40mg litre
.
1
1
.
1
.
phenoxy compounds were studied (Table 6). Good
activity was maintained even with large groups in the
3-position, as illustrated by 63 and 64. Substitution in
the 4-position was more limited. The 4-Ph and 4-tert-
Bu compounds 68 and 69 were inactive against Ph
infestans at 200mg litre 1, although the 4-Br com-
pound 66 was quite active. In summary, some of the
substituted compounds were nearly as active as
famoxadone, but none were signi®cantly more active
or possessed a desirable attribute which was lacking in
the unsubstituted material, and all of the substituted
compounds would have been more expensive to
prepare commercially.
3.3.3 Biological evaluation of the ring analogs
The biological activities for the 1-oxygen ring analogs
13 are shown in Table 8. Preventive control of Ph
infestans was lower for all three analogs 13a, 13b and
13c relative to the 2,4-oxazolidinedione (46), and
3.3.2 Biological activity of the enantiomers
The enantiomers of the lead compound (1) and
1
Electron transport IC50 (ngml
)
(ÆSEM)
Disease control of
1
a
Compound
Rat heart
Ph infestans
P viticola ED90 (mg ml
)
(R, S)-1
(R)-1
(S)-1
(R, S)-Famoxadone
(R)-Famoxadone
(S)-Famoxadone
45 (Æ3)b
1400 (Æ20)
17 (Æ2)
13 (Æ2)
450 (Æ14)
3.8 (Æ0.9)
Ð
10
80
3
4.5 (Æ1.4)c
3
200 (Æ11)
2.2 (Æ0.6)
Ð
10
0.4
Ð
a ED90 values were estimated from DuPont plant disease control reports. Dose rates used were 200, 40, 10,
Table 7. Selectivity of enantiomers of 1 and
famoxadone for inhibition of electron transport
in submitochondria and for greenhouse control
of Plasmopara viticola
1
2 and 0.4mg litre
.
b Mean and standard deviation from ®ve determinations.
c
Mean and standard deviation from 11 determinations.
Pest Manag Sci 57:143±152 (2001)
151

JA Sternberg et al
Picollelli MA, Schwartz RS, Sternberg JA and Tang X-S,
Oxazolidinones: a new chemical class of fungicides and inhibi-
tors of mitochondrial cyctochrome bc1 function. Pestic Sci
55:197±218 (1999).
there was no preventive control of P viticola at
40mg litre 1. However, other ring analogs illustrated
in Fig 10 possessed signi®cant fungicidal activity and
will be discussed at a future date. In fact, fenamidone
the emerging fungicide from Aventis, can be con-
sidered a ring analog of the 2,4-oxazolidinediones.
In summary, the structure-activity relationships
indicated that R1 and R3 were sensitive to structural
variation, whereas compounds with a wide variety of
R2 groups were fungicidally active. In the end, a 4-
phenoxyphenyl group was found to be optimal at this
position. This project ultimately led to the advance-
ment of famoxadone to commercial development for
the control of fungal infections of grapes, cereals,
tomatoes, potatoes, and other crops. Additional work
in the area, eg analogs with heterocyclic substituents
and process development studies,19 will be the subject
of future publications.
7 Geffken D, 3-Hydroxy-3, 3-alkoxy- und 3-amino-2-thioxoox-
azolidin-4-one aus 6-thioxoperhydro-1.5.2-dioxazin-3-onen. Z
Naturforsch, B: Anorg Chem, Org Chem 38B:1008±1014
(1983).
8 Geffken D, Synthese und eigenschaften von 4-benzyloxyimino-
1,3-dioxolan-2-onen(thionen). Chem Ztg 103:299±300 (1979).
9 Geffken D, 3-Alkoxyoxazolidin-2,4-dione aus N-alkoxy-2-hy-
droxycarbon-saureamiden und 1,1'-carbonyldiimidazol. Arch
Pharm (Weinheim, Germany) 313:817±825 (1980).
10 Geffken D, Synthese von 6,6-disubstituierten 2,5-dioxodihydro-
4H-1,3,4-dioxazinen und ihre umwandlung in 2-amino- bzw-
3-alkyl(aryl)-2,4-dioxotetrahydro-1,3-oxazole. Synthesis 1:38±
40 (1981).
11 Geffken D, Synthesis and properties of 4-unsubstituted and 4-
monosubstituted perhydro-1,5,2-dioxazine-3,6-diones. Arch
Pharm (Weinheim, Germany) 315:802±810 (1982).
12 Geffken D, 3-(1-Hydroxyalkyl)-1,4,2-dioxazol-5-ones and 3-
hydroxyoxazolidine-2,4-diones from 2-hydroxycarbohydroxa-
mic acids and 1,1'-carbonyldiimidazole. Liebigs Ann Chem
2:211±218 (1982).
ACKNOWLEDGEMENTS
13 Geffken D and Strohauer K, 4-Benzyloxy-2,3,5-morpholintrione
und 3-benzyloxy-2,4-oxazolidindione aus N-benzyloxyglyoko-
lamiden und oxalylchlorid. Arch Pharm (Weinheim, Germany)
319:577±582 (1986).
The authors would like to thank the many members of
Chemical Discovery who provided useful insights and
suggestions to the project, most notably Robert J
Pasteris and Kevin T Kranis, and the associate
scientists for their technical assistance. We also thank
John C Groce and Gina L Blankenship for technical
assistance in providing rapid, quality NMR analyses,
and the entire Plant Disease Control group for
biological evaluations.
14 Geffken D, Synthese und Eigenschaften von 3-alkoxy(hydroxy)-
thiazolidin-2,4-dionen. Z Naturforsch, B: Chem Sci 42b:1202±
1206 (1987).
15 Hauser M, The preparation and cyclization of chloroethyl carba-
zates. Some clari®cations. J Org Chem 31:968±970 (1966).
16 Kametani T, Cyclization reaction of N-substituted mandelhy-
drazide with formaldehyde. Heterocycles 9:1031±1040 (1978).
17 Hetzheim A and Mockel K, Recent advances in 1,3,4-oxadiazole
chemistry. Adv Heterocyclic Chem 7:183±224 (1966).
18 Geffken D, Hydrolytic desulfurization of some heterocyclic
thiones with silver nitrate. Arch Pharm (Weinheim, Germany)
312:363±365 (1979).
REFERENCES
1 Joshi MM and Sternberg JA, DPX-JE874: a broad-spectrum
fungicide with a new mode of action, in Proc 1996 Brighton Crop
Prot ConfÐPests and Diseases, British Crop Protection Council,
Farnham, UK, pp 21±26 (1996).
19 Sternberg JA, Sun K-M, Toji M and Witterholt VG, Imidazole
and triazole carboxylates, and processes for preparing 2,4-
oxazolidinediones, US Patent 5 552 554 (1996).
2 Geffken D and Rayner DR, Fungicidal oxazolidinones, US
Patent 4 957 933 (1990).
20 Sun K-M, New 5,5-Disubstituted 3-arylamino-imidazolin-4-one
derivatives, PCT Int Patent Appln WO 9 324 467 (1993).
21 Geffken D, Fungicidal iminooxazolidinones, PCT Int Patent
Appln WO 9 119 703 (1991).
3 Geffken D and Rayner DR, Fungicidal oxazolidinones, US
Patent 5 223 523 (1993).
4 Sternberg JA, Geffken D, Adams JB Jr , Jordan DB, PoÈstages R,
Sternberg CG, Campbell CL, Moberg WK and Livingston
RS, in Synthesis and Chemistry of Agrochemicals V, ed by Baker
DR, Feynes JG, Basarab GS and Hunt DA, American Chemi-
cal Society, Washington, DC, USA, pp 216±227 (1998).
5 Jordan DB, Livingston RS, Bisaha JJ, Duncan KE, Pember SO,
Picollelli MA, Schwartz RS, Sternberg JA and Tang X-S,
Mode of action of famoxadone. Pestic Sci 55:105±118 (1999).
6 Jordan DB, Livingston RS, Bisaha JJ, Duncan KE, Pember SO,
22 Adams JB Jr and Sternberg JA, New 4-thioxo-oxazolidin-2-one
and 4-imino-oxazolidin-2-one compounds, PCT Int Patent
Appln WO 9 216 515 (1992).
23 Campbell CL, New oxazolidinone derivatives, PCT Int Patent
Appln WO 9 322 299 (1993).
24 Campbell CL, Gross CM, Sternberg JA and Sun K-M, Novel
fungicidal oxazolidinones, PCT Int Patent Appln WO 9 318
016 (1993).
152
Pest Manag Sci 57:143±152 (2001)